Transitive Network (was Re: Network Ledger Technology: The First 15 Years)

23 views
Skip to first unread message

Michiel de Jong

unread,
Jan 10, 2019, 8:23:57 AM1/10/19
to Network Money
Hi Aniket,

Great project! I enjoyed reading your blogpost, and particulary like your use of the term 'Credit Network', it's probably a good term for a network where nodes represent economic agents and links represent credit relationships.

You say your goal is to look into the privacy acceptance. I agree that's a very important research topic. In particular, to make multi-hop ('transitive') transactions possible in a peer-to-peer network, some private information will almost inevitably have to leak to peers, unless the network is entirely anonymized like Lightning Network (but then the neighbor relationship is random, and that's not what we would generally understand as a credit network, where peers would normally be peers because they trust each other).

Sometimes I think that part of the reason people don't use p2p credit networks as much as  you might expect, is that they don't want risk their private credit relationships being broadcast across the network. Is that also the direction you're thinking in when you talk about privacy concerns in relation to credit networks?


Cheers,
Michiel.


On Wednesday, December 26, 2018 at 6:46:33 AM UTC+1, Aniket Kate wrote:
Thanks for the list. I enjoy this topic a lot, and believe that it need more attention.

My students and I have been looking into the privacy acceptance of credit networks. We also built a Solidity contract to understand it better. See our efforts at https://transitive.network/

I also wrote a more non-academic article in that direction at https://medium.com/@aniketkate/towards-a-tokenless-iou-credit-network-realization-in-ethereum-a512e11dcf5e

Looking for possible collaborations.

--Aniket

Prof. Aniket Kate
CS, Purdue University
https://www.cs.purdue.edu/homes/akate/

On Thursday, November 15, 2018 at 3:48:22 AM UTC-5, Michiel de Jong wrote:
Welcome to this mailinglist! With the title, "Network Money', I tried to keep the  subject broad, but centered around "money transfer without money movement", like we see in networks of ledgers, when they support multi-hop money transactions.

Please post your topics about projects you're working on, questions you run into, books you read, or interviews you enjoyed, and start a discussion!

My blogpost in which I announced the creation of this mailing list is here:


Enjoy, and don't forget to subscribe to this mailing list!

Cheers,
Michiel

cojulapa

unread,
Mar 11, 2019, 5:21:38 PM3/11/19
to Network Money
Hello Michiel,
finally I managed to sign up. :)
 
Sometimes I think that part of the reason people don't use p2p credit networks as much as  you might expect, is that they don't want risk their private credit relationships being broadcast across the network. Is that also the direction you're thinking in when you talk about privacy concerns in relation to credit networks?

I explored the same issue after diving into the topic of network moneys. 
For myself I defined 3 layers of money : 

Level 1: 
This is among close friends and family, most of the time. so people we trust a lot

People do things for each other, they might even lend money or pay stuff for each other, however, for most an accounting system would feel wierd here. It is like if you go out together and you pay the first round and you friend the second and the night goes on, but you are really counting with pen and paper the prices of the drinks to come out even... ;) 
This would be socially creapy and let you apear as cheeseparing or nitpicky. 
In my understanding at this level beeing even is really more a feeling, often also compensated by the fact that we like to have some people around us and doing them favours can be an honor to us. 
Therefore ,money is kind of invisible here, really just a concept of compensation graded by a feeling. (those who never pay a round or do a favour will probably not be our fiends too long)

Level 2: 
Are also as trusted contacts but more on a business level 
Here trust is based on a rational decision. We see them as reliable people or businesses.
Dragons Den would be of this kind, but coming from the crowd instead. 
Or if one has to track the status of IOU because it is business. An example of this might be if regular guests can chalk up their drinks. The shop owner allows this because it is clear anyway that they will come back, but the owner does it because in the end it must be clear that they are even and otherwise they would forget about the real numbers. 
There is clear risk but in the first example the incentive is the desire to help bring a product to market. In the second example the risk is taken because of inductive reasoning. 

Level 3:
Finally there a untrusted nodes which one still wants to pay. The simplest in social terms, you just want to buy the good they offer without making you dependent on each other.

As far as I can see, one should (mostly) forgert about level 1 first, when trying to get people to use network money. 
level 2 is a good starting point for introduction and enabling workable, expendient payment to Level 3 is then the tricky technical task. 

Reply all
Reply to author
Forward
0 new messages